US 6594538 B1 Resumen A method for identifying and characterizing shape variations in parts. Measurements are taken of a population of parts and the shapes of the parts are expressed as a function having a plurality of coefficients. Discrete or function error maps are developed from the coefficients and a principal components analysis is performed on the error maps to identify the principal components of variation of the parts. The parts may be grouped into sub-populations representing ranges of variation along each of the principal components of variation, and downstream processes may be controlled differently for each sub-population. In one embodiment, a typical (re-generated) part shape is identified along multiple principal components of variation, and a tool path is controlled to be responsive to the typical part shape. Information regarding the principal components of variation may further be used to revise upstream manufacturing processes to advantageously affect the distribution of error in subsequently manufactured parts.
Reclamaciones(22) 1. A method for identifying shape variations in parts, the method comprising the steps of:
obtaining measurements representing the shape of each of a plurality of parts;
using the measurements to express the shape of each part as a respective function, the function being one of the group of a discrete function, a function of a nominal part shape, and a function of a location in a part coordinate system, each function comprising a respective plurality of coefficients;
using the coefficients to define an error map for each respective part; and
performing a principal component analysis across the error maps to identify the principal components of variation in the error maps.
2. The method of
3. The method of
defining a plurality of ranges within at least one of the principal components of variation; and
defining a plurality of sub-populations wherein a part in a sub-population will exhibit a characteristic value of a principal component of variation within a respective one of the plurality of ranges.
4. The method of
obtaining a measurement representing the at least one of the principal components of variation for a subject part;
grouping the subject part into one of the sub-populations.
5. The method of
generating statistics showing the distribution of the plurality of parts among the plurality of sub-populations.
6. The method of
defining a plurality of ranges within each of a plurality of the principal components of variation; and
defining a plurality of sub-populations wherein a part in a sub-population will exhibit characteristic values for each of the plurality of principal components of variation within the respective plurality of ranges.
7. The method of
8. The method of
obtaining a measurement for the subject part at the at least one measurement location;
assigning the subject part to a sub-population; and
controlling a process utilizing the subject part in response to the assigned sub-population of the subject part.
9. The method of
defining a plurality of ranges within each of the principal components of variation; and
defining a plurality of sub-populations for each principal component of variation wherein a part in a sub-population will exhibit a respective principal component of variation within a respective one of the plurality of ranges.
10. The method of
grouping the plurality of parts into the respective sub-populations for each principal component of variation.
11. The method of
defining a plurality of ranges of the coefficients; and
defining a plurality of sub-populations wherein a part in a sub-population will exhibit coefficients within a respective one of the plurality of ranges.
12. The method of
obtaining measurements representing the shape of a subject part;
using the measurements to express the shape of the subject part as a function of the nominal part shape, the function comprising subject part coefficients; and
grouping the subject part into one of the sub-populations based upon which of the plurality of ranges into which a value of the subject part coefficients fall.
13. The method of
evaluating manufacturing processes used to produce the plurality of parts to identify a source of at least one of the principal components of variation; and
revising the manufacturing processes to effect the distribution of variation among parts produced by the manufacturing processes along the at least one of the principal components of variation.
14. The method of
15. The method of
defining a representative value of the variation of the plurality of parts for at least one principal component of variation; and
using the representative value to define a typical part shape.
16. The method of
using the typical part shape to define a tool path for a manufacturing process.
17. The method of
defining a representative value of the variation of the plurality of parts for each principal component of variation; and
using the representative values to define a typical part shape.
18. A method for identifying and characterizing shape variations in parts, the method comprising the steps of:
obtaining measurements representing the shape of each of a plurality of existing parts;
using the measurements to define an error map for each respective part, the error map representing the deviation of the part shape from a nominal part shape; and
performing a principal component analysis across the error maps to identify the principal components of variation in the error maps.
19. The method of
using the measurements to express the shape of each part as a respective function of a nominal part shape, each function comprising a respective plurality of coefficients; and
using the coefficients to define the error map for each respective part.
20. The method of
defining an error map along at least one of the principal components of variation at a predetermined level of statistical confidence;
constructing a geometric model from the error map defined along the at least one of the principal components of variation.
21. The method of
22. The method of
Descripción This application claims the benefit of the Dec. 10, 1999, filing date of provisional patent application 60/170,165. This invention relates generally to the field of manufacturing, and more particularly to the field of tolerance control for manufactured parts. Shape variation in manufactured parts is a normal occurrence. Parts are typically specified to have shapes that fall within predetermined dimensional limits. Dimensional control is important to downstream manufacturing operations because the success of some downstream manufacturing operations depends upon the as-manufactured shapes of the parts as they arrive to be processed. In a similar way, dimensional control is important to end users because it ensures the proper functionality of the part or final assembly. Furthermore, shape control is the key concept underlying the cost savings possible using interchangeable parts. In many cases, tolerances are assigned to ensure that the shape of the part will be predictable or repeatable. It is not the exact shape of the part that drives the tolerance band, rather, it is the level of required repeatability. In general, the greater the required repeatability, the smaller the tolerance bands, and the more expensive the part. There is a cost savings opportunity where the cost of holding very tight tolerances has outstripped the savings of using fully interchangeable parts. Thus there is a need for a method for controlling the variability of parts that does not rely solely on the control of manufacturing tolerances. With the convergence of flexible measuring technology in the form of Coordinate Measuring Machines (CMM's), and computer controlled Numerical Control (NC) machining, we have an opportunity to develop techniques of manufacture that do not rely on fully interchangeable parts, and thereby do not require the expense of extremely tight tolerances. The inventors have accomplish this by substituting shape predictability for shape repeatability. Accordingly, a method is disclosed herein for identifying shape variations in parts, the method comprising the steps of: obtaining measurements representing the shape of each of a plurality of parts; using the measurements to express the shape of each part as a respective function of a nominal part shape or as a function of a location in a part coordinate system, each function comprising a respective plurality of coefficients; and using the coefficients to define an error map for each respective part. Each respective function may be a discrete function or polynomial or trigonometric equation. The functions describing each of the plurality of parts can be combined to form a single function or set of functions describing the characteristics of the plurality. The method is further described as including the step of performing a principal component analysis across the error maps to identify the principal components of variation in the error maps. Once the principal components of variation are known, a plurality of sub-populations of the parts may be defined by identifying a plurality of ranges within at least one of the principal components of variation. Parts grouped into the various sub-populations may be treated differently for downstream processes. Furthermore, upstream processes may be controlled in order to affect the distribution of future parts within the various sub-populations. The features and advantages of the present invention will become apparent from the following detailed description of the invention when read with the accompanying drawings in which: FIG. 1 is a flow diagram of a process for identifying, characterizing and utilizing information regarding the shape of parts. FIG. 2 is a plan view of a nominal part. FIG. 3 illustrates a sample of parts measured with respect to the nominal part. FIG. 4 illustrates three values for each of three principal components of variation of the parts of FIG. FIG. 5 illustrates population frequency histograms for three principal components of variation. FIG. 6 illustrates a typical part compared to a nominal part. FIG. 7 illustrates the reduced number of measurement locations used for a subject part. FIG. 1 illustrates a method By measuring a plurality of similar parts and constructing error maps of their variations from nominal, it is possible to generate error maps that describe representative parts models which span the original distribution of measured parts. Such error maps describe various independent modes of variation. For purposes of this disclosure, an error map is a functional or discrete description of errors as they vary across the surface of a part. It includes such methods as discrete errors stored at individual locations, functional maps of error variation across the part surface, and maps of error variation as a function of spatial location. Methods of error mapping are well known and are not disclosed herein. A plurality of parts is manufactured at step An inspection plan is then created at step The next step The measurements may then be used in step Step
The coefficients may be derived from the functional equations developed in step In step
The principal components of variation in the error maps represent the independent patterns of variation within the parts. For example, FIG. 4 illustrates three principal components of variation It is possible at step The representative parts allow several interesting and novel applications in the manufacturing shop floor environment. For example, it is possible at step At step By using only those measurement locations that best correlate with overall modes of observed error, we can reduce the number of surface measurements required to disclose those errors. There is a savings in processing time to be realized by reducing the number of surface measurements, especially on parts with sculptured surfaces. Using these measurement locations The selection of measurement locations may also be made by considering the relative importance of the various components of variation. For example, the PCA performed to identify components of variation It is likely that parts grouped by principal components of variation will have different operating characteristics among the groups. Such characteristics may include differing part life, differing noise level during operation, differing efficiency of operation, etc. Therefore, it may be useful at step Knowledge of the principal components of variation may further be used at step It is possible at step In one embodiment of this invention, error maps may be constructed at step While the preferred embodiments of the present invention have been shown and described herein, it will be obvious that such embodiments are provided by way of example only. Numerous variations, changes and substitutions will occur to those of skill in the art without departing from the invention herein. Accordingly, it is intended that the invention be limited only by the spirit and scope of the appended claims. Citas de patentes
Citada por
Clasificaciones
Eventos legales
Girar |